Silicon-32, with a half-life of approximately 140 years, has the potential to fill the dating gap that lies between those chronologies based on the shorter-lived isotopes of 3H and 210Pb, and those based on the longer-lived 14C. Silicon-32 is produced in the atmosphere by cosmic ray bombardment of argon, and falls out on the Earth's surface in precipitation. Silicon-32 methods may be used to date siliceous sediments and sponges, groundwater and glacier ice. Measurement of 32Si concentrations in these archives is, however, not straightforward. Two methods are available: radioactive-decay counting of the activity of the daughter nucleus, 32P, and accelerator mass spectrometry, but in both cases the detection of 32Si pushes the boundaries of the technique. Even the half-life of 32Si is not known to a precision of better than ±10%. In this paper, we review efforts to determine the isotope's half-life, survey the detection methods and discuss the applications of 32Si chronology. We show that at least some of the chronometric potential of this radioisotope is close to being realised as a result of recent improvements in methods of measurement.
